
import os
import numpy as np
import pandas as pd
import seaborn as sns
import matplotlib.pyplot as plt
import pickle

from io import BytesIO
from zipfile import ZipFile
from urllib.request import urlopen

from sklearn.ensemble import RandomForestRegressor, GradientBoostingRegressor
from sklearn.feature_selection import SelectKBest, f_regression
from sklearn.linear_model import LinearRegression
from sklearn.metrics import mean_squared_error, r2_score
from sklearn.model_selection import GridSearchCV, train_test_split
from sklearn.preprocessing import StandardScaler
from sklearn.svm import LinearSVR
from sklearn.tree import DecisionTreeRegressor, plot_tree
from xgboost import XGBRegressor

# ensure openpyxl==3.1.0, version 3.1.1 has a bug


MajorRoadsEmissionsURL = 'https://data.london.gov.uk/download/london-atmospheric-emissions-inventory-2013/8c520de2-c518-4e64-932c-0071ac826742/LAEI2013_MajorRoads_EmissionsbyLink_2013.xlsx'
SupportingDocURL = 'https://data.london.gov.uk/download/london-atmospheric-emissions-inventory-2013/1f3755f3-dae8-4d39-b9b4-0cc7adb4e826/1.%20Supporting%20Information.zip'
LocalFileMajorRoadsEmissions = 'data/LAEI2013_MajorRoads_EmissionsbyLink_2013.xlsx'
LocalFileSupportingDoc = 'data/1. Supporting Information.zip'


# Check if data folder exists, if not create it
if not os.path.exists('data'):
    os.makedirs('data')

# Check if data is already downloaded, if not set load_from_local to False
if os.path.exists(LocalFileMajorRoadsEmissions):
    load_from_local = True
else:
    load_from_local = False

if load_from_local:
    MajorRoadsEmissionsURL = LocalFileMajorRoadsEmissions 
xls = pd.ExcelFile(MajorRoadsEmissionsURL)


if load_from_local:
    DocZip = ZipFile(LocalFileSupportingDoc)
else:
    resp = urlopen(SupportingDocURL)
    DocZip = ZipFile(BytesIO(resp.read()))
DocZip.namelist()[:10]

['1. Supporting Information/',
 '1. Supporting Information/1. Road Traffic Data/',
 '1. Supporting Information/1. Road Traffic Data/Excel/',
 '1. Supporting Information/1. Road Traffic Data/Excel/LAEI2013_2008_AADT-VKM.xlsx',
 '1. Supporting Information/1. Road Traffic Data/Excel/LAEI2013_2010_AADT-VKM.xlsx',
 '1. Supporting Information/1. Road Traffic Data/Excel/LAEI2013_2013_AADT-VKM.xlsx',
 '1. Supporting Information/1. Road Traffic Data/Excel/LAEI2013_2020_AADT-VKM.xlsx',
 '1. Supporting Information/1. Road Traffic Data/Excel/LAEI2013_2025_AADT-VKM.xlsx',
 '1. Supporting Information/1. Road Traffic Data/Excel/LAEI2013_2030_AADT-VKM.xlsx',
 '1. Supporting Information/1. Road Traffic Data/Excel/LAEI2013_VKM_AllYears_Summary.xlsx']


xls.sheet_names

['2013 LTS Rds', '2013 Other Major Rds']


df_LTSroads = pd.read_excel(xls, '2013 LTS Rds')
print(df_LTSroads.shape)
df_LTSroads.head()

(366220, 32)


df_OtherMajorRoads = pd.read_excel(xls, '2013 Other Major Rds')
print(df_OtherMajorRoads.shape)
df_OtherMajorRoads.head()

(513740, 32)


# concat  the two dataframes
df_RoadEmissions = pd.concat([df_LTSroads, df_OtherMajorRoads], ignore_index=True)
print(df_RoadEmissions.shape)
df_RoadEmissions.head()

(879960, 33)


traffic_excel = DocZip.read('1. Supporting Information/1. Road Traffic Data/Excel/LAEI2013_2013_AADT-VKM.xlsx')
traffic_xls = pd.ExcelFile(BytesIO(traffic_excel))


df_AADT = pd.read_excel(traffic_xls, 'MajorGrid_AADTandVKM_2013')
print(df_AADT.shape)
df_AADT.head()

(87999, 44)


# left join df_RoadEmissions and df_AADT on the ['Toid' and 'GRID_ExactCut_ID'] column
df = pd.merge(df_RoadEmissions, df_AADT, how='left',
            left_on=['Toid', 'GRID_ExactCut_ID'],
            right_on=['Toid', 'GRID_ExactCut_ID'])
print(df.shape)
df.head()

(879960, 75)


# save the joined dataset df to csv in data folder
df.to_csv('data/df.csv', index=False)


# remove unnecessary columns
columnstokeep = df.columns
for a in columnstokeep:
    if 'VKM' in a:
        columnstokeep = columnstokeep.drop(a)
df = df[columnstokeep]
print(df.shape)
df.head()

(879960, 58)


df.columns

Index(['GridId', 'Toid', 'GRID_ExactCut_ID', 'Location_ExactCut_x',
       'BoroughName_ExactCut_x', 'Lts', 'Length (m)_x', 'Emissions', 'Year_x',
       'Pollutant', 'Emissions Unit', 'Motorcycle', 'Taxi', 'Car',
       'BusAndCoach', 'Lgv', 'Rigid', 'Artic', 'Rigid2Axle', 'Rigid3Axle',
       'Rigid4Axle', 'Artic3Axle', 'Artic5Axle', 'Artic6Axle', 'PetrolCar',
       'DieselCar', 'PetrolLgv', 'DieselLgv', 'LtBus', 'Coach', 'ElectricCar',
       'ElectricLgv', 'DotRef', 'RowID', 'Year_y', 'Location_ExactCut_y',
       'BoroughName_ExactCut_y', 'TLRN', 'MotorwayNumber', 'AADT Motorcycle',
       'AADT Taxi', 'AADT Pcar', 'AADT Dcar', 'AADT PLgv', 'AADT DLgv',
       'AADT LtBus', 'AADT Coach', 'AADT Rigid2Axle', 'AADT Rigid3Axle',
       'AADT Rigid4Axle', 'AADT Artic3Axle', 'AADT Artic5Axle',
       'AADT Artic6Axle', 'AADT ElectricCar', 'AADT ElectricLgv', 'AADT TOTAL',
       'Speed (kph)', 'Length (m)_y'],
      dtype='object')


label = []
columnstokeep = df.columns
for a in columnstokeep:
    if 'AADT' in a:
        label = label + [a]
    elif 'Speed' in a:
        label = label + [a]
    elif 'Length' in a:
        label = label + [a]
label

['Length (m)_x',
 'AADT Motorcycle',
 'AADT Taxi',
 'AADT Pcar',
 'AADT Dcar',
 'AADT PLgv',
 'AADT DLgv',
 'AADT LtBus',
 'AADT Coach',
 'AADT Rigid2Axle',
 'AADT Rigid3Axle',
 'AADT Rigid4Axle',
 'AADT Artic3Axle',
 'AADT Artic5Axle',
 'AADT Artic6Axle',
 'AADT ElectricCar',
 'AADT ElectricLgv',
 'AADT TOTAL',
 'Speed (kph)',
 'Length (m)_y']


# remove irrelevant columns
label.remove('AADT TOTAL')
label.remove('Speed (kph)')
label.remove('Length (m)_y')
label

['Length (m)_x',
 'AADT Motorcycle',
 'AADT Taxi',
 'AADT Pcar',
 'AADT Dcar',
 'AADT PLgv',
 'AADT DLgv',
 'AADT LtBus',
 'AADT Coach',
 'AADT Rigid2Axle',
 'AADT Rigid3Axle',
 'AADT Rigid4Axle',
 'AADT Artic3Axle',
 'AADT Artic5Axle',
 'AADT Artic6Axle',
 'AADT ElectricCar',
 'AADT ElectricLgv']


target = ['Motorcycle', 'Taxi', 'Car',
          'BusAndCoach', 'Lgv', 'Rigid', 'Artic', 'Rigid2Axle', 'Rigid3Axle',
          'Rigid4Axle', 'Artic3Axle', 'Artic5Axle', 'Artic6Axle', 'PetrolCar',
          'DieselCar', 'PetrolLgv', 'DieselLgv', 'LtBus', 'Coach', 'ElectricCar',
          'ElectricLgv']


# save plot to visuals folder
if not os.path.exists('visuals'):
    os.makedirs('visuals')

# plot the correlation matrix
corr = df[label + target].corr()
plt.figure(figsize=(30, 30))
sns.heatmap(corr, annot=True, fmt='.2f', cmap='coolwarm')

plt.savefig('visuals/main_correlation_matrix.png')


# sum the target variables to one column
df['Total Emissions'] = df[target].sum(axis=1)


# Split df by unique 'Pollutant' values into dictionary
df_dict = {k: v for k, v in df.groupby('Pollutant')}
df_dict.keys()

dict_keys(['CO2', 'NOx', 'PM10_Brake', 'PM10_Exhaust', 'PM10_Resusp', 'PM10_Tyre', 'PM25_Brake', 'PM25_Exhaust', 'PM25_Resusp', 'PM25_Tyre'])


for k, v in df_dict.items():
    print(k, v.shape)

CO2 (87996, 59)
NOx (87996, 59)
PM10_Brake (87996, 59)
PM10_Exhaust (87996, 59)
PM10_Resusp (87996, 59)
PM10_Tyre (87996, 59)
PM25_Brake (87996, 59)
PM25_Exhaust (87996, 59)
PM25_Resusp (87996, 59)
PM25_Tyre (87996, 59)


from sklearn.model_selection import train_test_split

train_test_set = {}

for k, v in df_dict.items():
    x_train, x_test, y_train, y_test = train_test_split(v[label], v['Total Emissions'], test_size=0.2, random_state=42)
    # drop indexes
    x_train = x_train.reset_index(drop=True)
    x_test = x_test.reset_index(drop=True)
    y_train = y_train.reset_index(drop=True)
    y_test = y_test.reset_index(drop=True)
    train_test_set[k] = {'x_train': x_train, 'x_test': x_test, 'y_train': y_train, 'y_test': y_test}

train_test_set.keys()

dict_keys(['CO2', 'NOx', 'PM10_Brake', 'PM10_Exhaust', 'PM10_Resusp', 'PM10_Tyre', 'PM25_Brake', 'PM25_Exhaust', 'PM25_Resusp', 'PM25_Tyre'])


# feature selection
for k, v in train_test_set.items():
    selector = SelectKBest(f_regression, k=10)
    selector.fit(v['x_train'], v['y_train'])
    v['x_train'] = selector.transform(v['x_train'])
    v['x_test'] = selector.transform(v['x_test']) 
    v['selected_features'] = [label[i] for i in selector.get_support(indices=True)]
    v['x_train'] = pd.DataFrame(v['x_train'], columns=v['selected_features'])
    v['x_test'] = pd.DataFrame(v['x_test'], columns=v['selected_features'])
    v['selector'] = selector
    print(k, v['selected_features'])

CO2 ['Length (m)_x', 'AADT Pcar', 'AADT Dcar', 'AADT PLgv', 'AADT DLgv', 'AADT Rigid2Axle', 'AADT Rigid3Axle', 'AADT Artic3Axle', 'AADT Artic5Axle', 'AADT Artic6Axle']
NOx ['Length (m)_x', 'AADT Pcar', 'AADT Dcar', 'AADT PLgv', 'AADT DLgv', 'AADT Rigid2Axle', 'AADT Rigid3Axle', 'AADT Artic3Axle', 'AADT Artic5Axle', 'AADT Artic6Axle']
PM10_Brake ['Length (m)_x', 'AADT Pcar', 'AADT Dcar', 'AADT PLgv', 'AADT DLgv', 'AADT Rigid2Axle', 'AADT Rigid3Axle', 'AADT Artic3Axle', 'AADT Artic5Axle', 'AADT Artic6Axle']
PM10_Exhaust ['Length (m)_x', 'AADT Pcar', 'AADT Dcar', 'AADT PLgv', 'AADT DLgv', 'AADT Rigid2Axle', 'AADT Rigid3Axle', 'AADT Artic3Axle', 'AADT Artic5Axle', 'AADT Artic6Axle']
PM10_Resusp ['Length (m)_x', 'AADT Pcar', 'AADT Dcar', 'AADT PLgv', 'AADT DLgv', 'AADT Rigid2Axle', 'AADT Rigid3Axle', 'AADT Artic3Axle', 'AADT Artic5Axle', 'AADT Artic6Axle']
PM10_Tyre ['Length (m)_x', 'AADT Pcar', 'AADT Dcar', 'AADT PLgv', 'AADT DLgv', 'AADT Rigid2Axle', 'AADT Rigid3Axle', 'AADT Artic3Axle', 'AADT Artic5Axle', 'AADT Artic6Axle']
PM25_Brake ['Length (m)_x', 'AADT Pcar', 'AADT Dcar', 'AADT PLgv', 'AADT DLgv', 'AADT Rigid2Axle', 'AADT Rigid3Axle', 'AADT Artic3Axle', 'AADT Artic5Axle', 'AADT Artic6Axle']
PM25_Exhaust ['Length (m)_x', 'AADT Pcar', 'AADT Dcar', 'AADT PLgv', 'AADT DLgv', 'AADT Rigid2Axle', 'AADT Rigid3Axle', 'AADT Artic3Axle', 'AADT Artic5Axle', 'AADT Artic6Axle']
PM25_Resusp ['Length (m)_x', 'AADT Pcar', 'AADT Dcar', 'AADT PLgv', 'AADT DLgv', 'AADT Rigid2Axle', 'AADT Rigid3Axle', 'AADT Artic3Axle', 'AADT Artic5Axle', 'AADT Artic6Axle']
PM25_Tyre ['Length (m)_x', 'AADT Pcar', 'AADT Dcar', 'AADT PLgv', 'AADT DLgv', 'AADT Rigid2Axle', 'AADT Rigid3Axle', 'AADT Artic3Axle', 'AADT Artic5Axle', 'AADT Artic6Axle']


# create a table using dataframe for the selected features
df_selected_features = pd.DataFrame(columns=['Pollutant', 'Feature', 'Score', 'Pvalue'])
for k, v in train_test_set.items():
    for i in range(len(v['selected_features'])):
        # using concat to append a row to the dataframe
        df_selected_features = pd.concat([df_selected_features,
                                            pd.DataFrame([[k, v['selected_features'][i],
                                                              v['selector'].scores_[i],
                                                                v['selector'].pvalues_[i]]],
                                                            columns=['Pollutant', 'Feature', 'Score', 'Pvalue'])],
                                            ignore_index=True)
df_selected_features = df_selected_features.sort_values(by=['Pollutant', 'Score'], ascending=False)
df_selected_features.head()


# create output folder if it does not exist
if not os.path.exists('features'):
    os.makedirs('features')

# save df_selected_features to csv in features folder
df_selected_features.to_csv('features/df_selected_features.csv', index=False)


# create output folder if it does not exist
if not os.path.exists('output'):
    os.makedirs('output')

# output the training and testing sets to csv files in the output folder, with the target on the left
for k, v in train_test_set.items():
    df_train_output = pd.concat([v['y_train'], v['x_train']], axis=1)
    df_test_output = pd.concat([v['y_test'], v['x_test']], axis=1)
    df_train_output.to_csv('output/' + k + '_train.csv', index=False)
    df_test_output.to_csv('output/' + k + '_test.csv', index=False)


# visualize correlation matrix of selected features
for k, v in train_test_set.items():
    corr = pd.DataFrame(v['x_train'], columns=v['selected_features']).corr()
    # set title of plt to k
    print(k)
    plt.figure(figsize=(10,10))
    sns.heatmap(corr, annot=True, fmt='.2f', cmap='coolwarm')
    plt.savefig('visuals/' + k + '_correlation_matrix.png')

CO2
NOx
PM10_Brake
PM10_Exhaust
PM10_Resusp
PM10_Tyre
PM25_Brake
PM25_Exhaust
PM25_Resusp
PM25_Tyre


# if train_test_set is not initialized, load the csv files in the output folder
try:
    train_test_set
except NameError:
    train_test_set = {}
    for file in os.listdir('output'):
        if file.endswith('_train.csv'):
            # exclude _train.csv from file name
            pollutant = file[:-len('_train.csv')]
            df_train = pd.read_csv('output/' + file)
            df_test = pd.read_csv('output/' + pollutant + '_test.csv')
            train_test_set[pollutant] = {'x_train': df_train.iloc[:, 1:],
                                         'x_test': df_test.iloc[:, 1:],
                                         'y_train': df_train.iloc[:, 0],
                                         'y_test': df_test.iloc[:, 0]}
            train_test_set[pollutant]['selected_features'] = list(train_test_set[pollutant]['x_train'].columns)


# stardaize the x_train and x_test
for k, v in train_test_set.items():
    # initialize the scaler
    scaler = StandardScaler()
    # fit the scaler to the x_train
    scaler.fit(v['x_train'])
    # transform the x_train and x_test
    v['x_train'] = scaler.transform(v['x_train'])
    v['x_test'] = scaler.transform(v['x_test'])
    v['x_train'] = pd.DataFrame(v['x_train'], columns=v['selected_features'])
    v['x_test'] = pd.DataFrame(v['x_test'], columns=v['selected_features'])
    # save the scaler
    v['scaler'] = scaler


model_dict = {}


def train_eval_model(model, x_train, y_train, x_test, y_test):
    # train the model
    model.fit(x_train, y_train)
    # predict the y_test
    y_pred = model.predict(x_test)
    # evaluate the model
    mse = mean_squared_error(y_test, y_pred)
    rmse = np.sqrt(mse)
    r2 = r2_score(y_test, y_pred)
    return {'model': model, 'mse': mse, 'rmse': rmse, 'r2': r2}


def hyperparameter_tuning(model, param_grid, pollutant, model_dict, x_train, y_train, x_test, y_test):
    # initialize the grid search
    grid_search = GridSearchCV(model, param_grid, cv=5)
    # fit the grid search to the training data
    grid_search.fit(x_train, y_train)
    # save the best model
    model_dict[pollutant]['best_model'] = grid_search.best_estimator_
    # predict the y_test
    y_pred = grid_search.best_estimator_.predict(x_test)
    # evaluate the model
    mse = mean_squared_error(y_test, y_pred)
    rmse = np.sqrt(mse)
    r2 = r2_score(y_test, y_pred)
    model_dict[pollutant]['best_mse'] = mse
    model_dict[pollutant]['best_rmse'] = rmse
    model_dict[pollutant]['best_r2'] = r2
    model_dict[pollutant]['best_params'] = grid_search.best_params_
    model_dict[pollutant]['best_score'] = grid_search.best_score_
    return model_dict


# model dict for each pollutant
LR_dict = {}

for k, v in train_test_set.items():
    LR_dict[k] = train_eval_model(
        LinearRegression(), v['x_train'], v['y_train'], v['x_test'], v['y_test'])
    print(k, LR_dict[k]['mse'], LR_dict[k]['r2'], LR_dict[k]['rmse'])

CO2 156343.23707624507 0.6228061024706681 395.40262654191497
NOx 1.3337240174376739 0.5887914721059816 1.1548696971683317
PM10_Brake 0.002064996161808893 0.581905620953646 0.04544222883848121
PM10_Exhaust 0.0008067369355814339 0.5964592889324114 0.028403114892233808
PM10_Resusp 0.007830744110917233 0.5904504796613461 0.08849149174308925
PM10_Tyre 0.00028900442231542513 0.650662925495141 0.017000130067603165
PM25_Brake 0.00035156196710148805 0.566541695180498 0.01874998578936763
PM25_Exhaust 0.0005769161710396386 0.5835788811077669 0.024019079312905367
PM25_Resusp 1.0696322799024033e-05 0.5908667099465907 0.0032705233218896378
PM25_Tyre 0.00015482685366526594 0.6211541868187214 0.012442943930809379


# add LR_dict to model_dict
model_dict['LR'] = LR_dict


# model dict for each pollutant
SVR_dict = {}

for k, v in train_test_set.items():
    SVR_dict[k] = train_eval_model(LinearSVR(random_state=42), v['x_train'], v['y_train'], v['x_test'], v['y_test'])
    print(k, SVR_dict[k]['mse'], SVR_dict[k]['r2'], SVR_dict[k]['rmse'])

CO2 206120.64719294346 0.5027130579491648 454.0051180250543

c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(

NOx 1.550136473433082 0.522068037359138 1.2450447676421448

c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(

PM10_Brake 0.0023693446396178194 0.5202849796194939 0.048675914368585

c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(

PM10_Exhaust 0.0009373092222723734 0.5311452676040003 0.030615506239034713

c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(

PM10_Resusp 0.00965171057585185 0.4952135607040501 0.09824311973798394

c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(


param_grid = {'C': [0.1, 1, 10, 100],
              'epsilon': [0.01, 0.001, 0.0001]}
for k, v in train_test_set.items():
    SVR_dict = hyperparameter_tuning(LinearSVR(random_state=42, max_iter=2000), param_grid, k, SVR_dict, v['x_train'], v['y_train'], v['x_test'], v['y_test'])
    print(k, SVR_dict[k]['best_params'], SVR_dict[k]['best_mse'], SVR_dict[k]['best_r2'], SVR_dict[k]['best_rmse'])

CO2 {'C': 100, 'epsilon': 0.001} 193877.50460409 0.5322508797152887 440.3152332183046

c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(

NOx {'C': 10, 'epsilon': 0.001} 1.5790238442882274 0.5131615971295512 1.2565921551116843

c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(

PM10_Brake {'C': 1, 'epsilon': 0.01} 0.0024926437290882304 0.49532093503552965 0.049926383096397346

c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(

PM10_Exhaust {'C': 0.1, 'epsilon': 0.01} 0.0009492061246991035 0.5251942763291007 0.030809188965292538

c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(

PM10_Resusp {'C': 100, 'epsilon': 0.01} 0.009010656116329575 0.5287408401923784 0.09492447585491097

c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(
c:\Users\zhu\miniconda3\lib\site-packages\sklearn\svm\_base.py:1244: ConvergenceWarning: Liblinear failed to converge, increase the number of iterations.
  warnings.warn(


model_dict['SVR'] = SVR_dict


# model dict for each pollutant
DTR_dict = {}

for k, v in train_test_set.items():
    DTR_dict[k] = train_eval_model(DecisionTreeRegressor(max_depth=5), v['x_train'], v['y_train'], v['x_test'], v['y_test'])
    # add x_train to DTR_dict
    DTR_dict[k]['x_train'] = v['x_train']
    print(k, DTR_dict[k]['mse'], DTR_dict[k]['r2'], DTR_dict[k]['rmse'])

CO2 26796.950698261982 0.9353496418214142 163.69774188504246
NOx 0.2781550911232453 0.9142403195476974 0.5274041060925155
PM10_Brake 0.0011160096951484224 0.7740444320758151 0.03340673128500336
PM10_Exhaust 0.00012539893120846476 0.9372737609559078 0.011198166421716761
PM10_Resusp 0.0009744026984635491 0.9490385393622965 0.031215424047472896
PM10_Tyre 5.022991487854874e-05 0.9392840726251969 0.007087306602549994
PM25_Brake 0.0002281752809426322 0.7186713019768846 0.015105471887452976
PM25_Exhaust 9.613658983853143e-05 0.9306081050997367 0.009804926814542342
PM25_Resusp 1.4630324990932853e-06 0.9440391514863675 0.001209558803487158
PM25_Tyre 2.895337285167896e-05 0.9291539947840726 0.005380833843530105


for k, v in DTR_dict.items():
    fig, ax = plt.subplots(figsize=(120, 12))
    plot_tree(v['model'],
            ax=ax,
            feature_names=v['x_train'].columns,
            fontsize=12,
            filled=True)
    plt.savefig('visuals/' + k + '_decision_tree.png')


param_grid = {'max_depth': [2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, None]}
for k, v in train_test_set.items():
    DTR_dict = hyperparameter_tuning(DecisionTreeRegressor(), param_grid, k, DTR_dict, v['x_train'], v['y_train'], v['x_test'], v['y_test'])
    print(k, DTR_dict[k]['best_params'], DTR_dict[k]['best_score'], DTR_dict[k]['best_mse'], DTR_dict[k]['best_r2'], DTR_dict[k]['best_rmse'])

CO2 {'max_depth': 12} 0.9697240433391443 10141.367018728306 0.9755329247135643 100.70435451721194
NOx {'max_depth': 11} 0.9532745244751236 0.1533773723209562 0.9527112935961652 0.3916342328256765
PM10_Brake {'max_depth': 9} 0.7762390474883176 0.0009654341375156992 0.8045310719216077 0.031071436038839582
PM10_Exhaust {'max_depth': 12} 0.970536735750741 6.33419791509477e-05 0.9683154865240187 0.007958767439179745
PM10_Resusp {'max_depth': None} 0.9793104888627939 0.000277300536009141 0.9854971251896912 0.01665234325880718
PM10_Tyre {'max_depth': 11} 0.9728880194342917 1.9301499305373974e-05 0.976669113756544 0.0043933471642215996
PM25_Brake {'max_depth': 9} 0.7852800279039471 0.00014973459928669048 0.8153847354002886 0.012236608978254166
PM25_Exhaust {'max_depth': 12} 0.9716421510864082 4.403156426432581e-05 0.9682177859142277 0.006635628400108448
PM25_Resusp {'max_depth': 12} 0.9780218511170398 4.516586770220282e-07 0.9827241002367605 0.0006720555609635473
PM25_Tyre {'max_depth': None} 0.9730447966223146 8.00454787131006e-06 0.9804136725918933 0.0028292309681802333


model_dict['DTR'] = DTR_dict


RFR_dict = {}

for k, v in train_test_set.items():
    RFR_dict[k] = train_eval_model(RandomForestRegressor(random_state=42), v['x_train'], v['y_train'], v['x_test'], v['y_test'])
    print(k, RFR_dict[k]['mse'], RFR_dict[k]['r2'], RFR_dict[k]['rmse'])

CO2 5182.412786670413 0.9874969041567384 71.98897684139158
NOx 0.0793740868329035 0.9755276946559032 0.28173407112542054
PM10_Brake 0.0006430227269420954 0.8698088680716739 0.02535789279380476
PM10_Exhaust 2.9690074433202608e-05 0.9851486237706606 0.0054488599204973705
PM10_Resusp 0.00012833779545375515 0.9932879070207226 0.011328627253721218
PM10_Tyre 1.0223574497084213e-05 0.9876421489429809 0.0031974324851487034
PM25_Brake 0.00010593440864047174 0.8693881776520503 0.010292444250054102
PM25_Exhaust 2.619579850298466e-05 0.9810917352113209 0.0051181831251904866
PM25_Resusp 1.9976005031682747e-07 0.9923591978156442 0.0004469452430855792
PM25_Tyre 4.6307604902270805e-06 0.9886689926066657 0.0021519201867697324


param_grid = {'n_estimators': [120, 150, 200]}
for k, v in train_test_set.items():
    RFR_dict = hyperparameter_tuning(RandomForestRegressor(random_state=42), param_grid, k, RFR_dict, v['x_train'], v['y_train'], v['x_test'], v['y_test'])
    print(k, RFR_dict[k]['best_params'], RFR_dict[k]['best_mse'], RFR_dict[k]['best_r2'], RFR_dict[k]['best_rmse'])

CO2 {'n_estimators': 200} 5181.916028565403 0.98749810263599 71.98552652141542
NOx {'n_estimators': 200} 0.07917984362339837 0.9755875829560235 0.2813891320278706
PM10_Brake {'n_estimators': 120} 0.0006479354559862355 0.86881420065431 0.02545457632698363
PM10_Exhaust {'n_estimators': 150} 3.0021769435629244e-05 0.9849827054505327 0.005479212483161175
PM10_Resusp {'n_estimators': 150} 0.00012748472250863533 0.9933325229104182 0.011290913271681585
PM10_Tyre {'n_estimators': 200} 9.870728996230191e-06 0.9880686545792372 0.0031417716333671025
PM25_Brake {'n_estimators': 200} 0.00010477083756667516 0.8708228025329586 0.010235762676355639
PM25_Exhaust {'n_estimators': 200} 2.6032444649002596e-05 0.9812096448801173 0.005102199981282838
PM25_Resusp {'n_estimators': 200} 1.982879971088954e-07 0.9924155037053795 0.0004452954043204302
PM25_Tyre {'n_estimators': 200} 4.686965799321414e-06 0.9885314638413071 0.0021649401375838117


model_dict['RFR'] = RFR_dict


GBR_dict = {}

for k, v in train_test_set.items():
    GBR_dict[k] = train_eval_model(GradientBoostingRegressor(random_state=42), v['x_train'], v['y_train'], v['x_test'], v['y_test'])
    print(k, GBR_dict[k]['mse'], GBR_dict[k]['r2'], GBR_dict[k]['rmse'])

CO2 8633.984438343794 0.9791696378915432 92.91923610503798
NOx 0.12507121446808328 0.9614385365256115 0.3536540887195895
PM10_Brake 0.0007463028495536954 0.8488980114174403 0.02731854405991826
PM10_Exhaust 5.23467732373466e-05 0.9738154370246256 0.007235106995570045
PM10_Resusp 0.00028455085172488724 0.9851179321932622 0.016868635147067684
PM10_Tyre 1.6312135086506545e-05 0.9802825385701923 0.00403882842994185
PM25_Brake 0.00012991753206301817 0.8398181871643908 0.011398137218994082
PM25_Exhaust 3.9747194771496814e-05 0.9713102662909461 0.0063045376334428214
PM25_Resusp 3.9588097782204745e-07 0.9848575917192146 0.0006291907324667517
PM25_Tyre 8.118895141196143e-06 0.9801338762683258 0.00284936749844525


# grid search for best parameters for GradientBoostingRegressor
param_grid = {'n_estimators': [120, 150, 200],
                'max_depth': [4, 6]}
for k, v in train_test_set.items():
    GBR_dict = hyperparameter_tuning(GradientBoostingRegressor(random_state=42),
                                    param_grid,
                                    pollutant=k,
                                    model_dict=GBR_dict,
                                    x_train=v['x_train'],
                                    y_train=v['y_train'],
                                    x_test=v['x_test'],
                                    y_test=v['y_test'])
    print(k, GBR_dict[k]['best_params'], GBR_dict[k]['best_mse'], GBR_dict[k]['best_r2'], GBR_dict[k]['best_rmse'])

CO2 {'max_depth': 6, 'n_estimators': 200} 4596.919651769342 0.9889094656570628 67.80058739988424
NOx {'max_depth': 6, 'n_estimators': 200} 0.07728462765420593 0.9761719084675563 0.2780011288721791
PM10_Brake {'max_depth': 4, 'n_estimators': 200} 0.0006247586432570933 0.8735067493890137 0.024995172399027245
PM10_Exhaust {'max_depth': 6, 'n_estimators': 200} 2.5179368362380418e-05 0.9874049398694451 0.0050179047781300536
PM10_Resusp {'max_depth': 6, 'n_estimators': 200} 0.00011775479221586277 0.9938414002569246 0.010851488018509848
PM10_Tyre {'max_depth': 6, 'n_estimators': 200} 8.046847036130969e-06 0.9902732906989152 0.0028366965005320833
PM25_Brake {'max_depth': 4, 'n_estimators': 200} 0.00010093230578376914 0.8755555200488266 0.010046507143468776
PM25_Exhaust {'max_depth': 6, 'n_estimators': 200} 2.0325509738706752e-05 0.9853289404382709 0.004508382164225517
PM25_Resusp {'max_depth': 6, 'n_estimators': 200} 1.7520702533832494e-07 0.9932983485947451 0.000418577382736245
PM25_Tyre {'max_depth': 6, 'n_estimators': 200} 4.205481540390601e-06 0.989709607627675 0.0020507270760368383


model_dict['GBR'] = GBR_dict


XGB_dict = {}

for k, v in train_test_set.items():
    XGB_dict[k] = train_eval_model(XGBRegressor(random_state=42), v['x_train'], v['y_train'], v['x_test'], v['y_test'])
    print(k, XGB_dict[k]['mse'], XGB_dict[k]['r2'], XGB_dict[k]['rmse'])

CO2 4248.047338888423 0.9897511554537981 65.1770461043489
NOx 0.07502231582260359 0.9768694155272984 0.27390201865375796
PM10_Brake 0.0006517468906207693 0.8680425094394434 0.025529333924346115
PM10_Exhaust 2.649124024145316e-05 0.9867487238372276 0.0051469641772071
PM10_Resusp 0.00012123851173245462 0.9936592010129184 0.011010836105058264
PM10_Tyre 8.809767276381068e-06 0.989351102994401 0.002968125212382569
PM25_Brake 0.00010614136977098437 0.869133004939441 0.010302493376410603
PM25_Exhaust 2.22905163790261e-05 0.9839105883364085 0.004721283340261003
PM25_Resusp 2.363324085222132e-07 0.9909603087283639 0.0004861403177295761
PM25_Tyre 4.437779057876622e-06 0.9891411988547224 0.002106603678406696


# grid search for best parameters for XGBRegressor
param_grid = {'n_estimators': [120, 150, 200],
                'max_depth': [6, 8]}
for k, v in train_test_set.items():
    XGB_dict = hyperparameter_tuning(XGBRegressor(random_state=42),
                                        param_grid,
                                        pollutant=k,
                                        model_dict=XGB_dict,
                                        x_train=v['x_train'],
                                        y_train=v['y_train'],
                                        x_test=v['x_test'],
                                        y_test=v['y_test'])
    print(k, XGB_dict[k]['best_params'], XGB_dict[k]['best_mse'], XGB_dict[k]['best_r2'], XGB_dict[k]['best_rmse'])

CO2 {'max_depth': 6, 'n_estimators': 200} 3726.5828714324016 0.9910092413075973 61.045744089431835
NOx {'max_depth': 8, 'n_estimators': 200} 0.06234448170662684 0.9807781953314824 0.24968876968463527
PM10_Brake {'max_depth': 6, 'n_estimators': 120} 0.0006546082734722817 0.8674631727275199 0.025585313628569844
PM10_Exhaust {'max_depth': 6, 'n_estimators': 200} 2.4332260875297865e-05 0.9878286744756233 0.004932774156121266
PM10_Resusp {'max_depth': 6, 'n_estimators': 200} 0.00010098858053788857 0.9947182765605491 0.0100493074655863
PM10_Tyre {'max_depth': 6, 'n_estimators': 200} 7.934530900636035e-06 0.9904090539854391 0.0028168299381815783
PM25_Brake {'max_depth': 6, 'n_estimators': 120} 0.00010611452254087317 0.869166106230087 0.010301190345822815
PM25_Exhaust {'max_depth': 6, 'n_estimators': 200} 2.028743938985613e-05 0.9853564198157971 0.0045041580111998885
PM25_Resusp {'max_depth': 8, 'n_estimators': 120} 2.2276216625497223e-07 0.9914793691540763 0.0004719768704660984
PM25_Tyre {'max_depth': 6, 'n_estimators': 200} 3.98262336729109e-06 0.9902549192697659 0.001995651113619585


model_dict['XGB'] = XGB_dict


if not os.path.exists('pickles'):
    os.makedirs('pickles')

if not os.path.exists('metrics'):
    os.makedirs('metrics')


# Save the model_dict to a pickle file
with open('pickles/model_dict.pkl', 'wb') as f:
    pickle.dump(model_dict, f)


# consolidate the metrics for each model from model_dict to a csv file
df_metrics = pd.DataFrame()
columns_list = ['model_name', 'pollutant', 'mse', 'rmse', 'r2', 'best_mse', 'best_rmse', 'best_r2', 'best_params']
for k, v in model_dict.items():
    for k1, v1 in v.items():
        if 'best_model' in v1.keys():
            row = [k, k1, v1['mse'], v1['rmse'], v1['r2'],
                    v1['best_mse'], v1['best_rmse'], v1['best_r2'], v1['best_params']]
        else:
            row = [k, k1, v1['mse'], v1['rmse'], v1['r2'], np.nan, np.nan, np.nan, np.nan]
        df_metrics = pd.concat([df_metrics, pd.DataFrame([row], columns=columns_list)], axis=0)

# copy mse, rmse and r2 from LR in df_metrics to best_mse, best_rmse and best_r2
df_metrics.loc[df_metrics['model_name'] == 'LR', 'best_mse'] = df_metrics.loc[df_metrics['model_name'] == 'LR', 'mse']
df_metrics.loc[df_metrics['model_name'] == 'LR', 'best_rmse'] = df_metrics.loc[df_metrics['model_name'] == 'LR', 'rmse']
df_metrics.loc[df_metrics['model_name'] == 'LR', 'best_r2'] = df_metrics.loc[df_metrics['model_name'] == 'LR', 'r2']
df_metrics.loc[df_metrics['model_name'] == 'LR', 'best_params'] = 'default'

df_metrics.to_csv('metrics/consolidated_metrics.csv')


# if df_selected_features is not initialized, load from features folder
try:
    df_selected_features
except NameError:
    df_selected_features = pd.read_csv('features/selected_features.csv')

df_selected_features.head()


# if model_dict is not initialized, load from models folder
try:
    model_dict
except NameError:
    with open('pickles/model_dict.pkl', 'rb') as f:
        model_dict = pickle.load(f)

# if df_metrics is not initialized, load from metrics folder
try:
    df_metrics
except NameError:
    df_metrics = pd.read_csv('metrics/consolidated_metrics.csv')

df_metrics.head()


def get_best_model(df, best_metric, min_max):
    """
    Get the best model for each pollutant based on the best_metric
    """
    # Create a pivot table with pollutant as the index, model_name as the columns, and best_mse as the values
    pivot = pd.pivot_table(df, index='pollutant',
                           columns='model_name', values=best_metric)

    # get the lowest or highest value for each row, the corresponding model_name for each column
    if min_max == 'min':
        best_model = pivot.idxmin(axis=1)
    elif min_max == 'max':
        best_model = pivot.idxmax(axis=1)

    column_names = ['pollutant', best_metric]
    df_best_metric = pd.DataFrame(columns=column_names)

    # convert best_model to df_best_mse
    for i in range(len(best_model)):
        df_best_metric = pd.concat([df_best_metric, pd.DataFrame([[best_model.index[i], best_model[i]]], columns=column_names)], axis=0)

    return df_best_metric


df_best_r2 = get_best_model(df_metrics, 'best_r2', 'max')
df_best_mse = get_best_model(df_metrics, 'best_mse', 'min')
df_best_rmse = get_best_model(df_metrics, 'best_rmse', 'min')

# combine df_best_r2, df_best_mse and df_best_rmse
df_best_model = pd.merge(df_best_r2, df_best_mse, on='pollutant')
df_best_model = pd.merge(df_best_model, df_best_rmse, on='pollutant')

df_best_model.to_csv('metrics/best_model.csv')

df_best_model


# create and save pivot table for each pollutant, pivot based on model_name and the metrics
for k, v in df_metrics.groupby('pollutant'):
    pivot = pd.pivot_table(v, index='model_name', values=['mse', 'rmse', 'r2', 'best_mse', 'best_rmse', 'best_r2'])
    pivot.to_csv('metrics/pivot_{}.csv'.format(k))


# create and save a table with the model_name as index, a row with the best_params for each model, a column for each pollutant
df_best_params = pd.DataFrame(index=df_metrics['model_name'].unique())
for k, v in df_metrics.groupby('pollutant'):
    df_best_params[k] = v['best_params'].values

# transpose the df_best_params
df_best_params = df_best_params.T

df_best_params.to_csv('metrics/best_params.csv')


# average the best_mse, best_rmse and best_r2 for each model across all pollutants
df_average_metrics = df_metrics.groupby('model_name').mean(numeric_only=True)[['best_mse', 'best_rmse', 'best_r2']]
df_average_metrics.to_csv('metrics/average_metrics.csv')

	GridId	Toid	GRID_ExactCut_ID	Location_ExactCut	BoroughName_ExactCut	Lts	Length (m)	Emissions	Year	Pollutant	...	Artic5Axle	Artic6Axle	PetrolCar	DieselCar	PetrolLgv	DieselLgv
0	6253	4000000027908919	24	External	NonGLA	18898	50.761449	DFT	2013	CO2	...	0.241372	0.190560	8.761443	4.810774	0.037550	1.735121
1	6253	4000000027947931	24	External	NonGLA	18895	28.592125	DFT	2013	CO2	...	0.000000	0.000000	0.015535	0.008576	0.000000	0.000000
2	6253	4000000028013383	24	External	NonGLA	15816	5.101391	DFT	2013	CO2	...	0.027271	0.021509	0.939028	0.518684	0.004055	0.184415
3	6253	4000000028025820	24	External	NonGLA	15816	3.757501	DFT	2013	CO2	...	0.020087	0.015843	0.691654	0.382044	0.002987	0.135834
4	6253	4000000028029388	24	External	NonGLA	15816	1.624593	DFT	2013	CO2	...	0.008685	0.006850	0.299044	0.165180	0.001292	0.058729

	GridId	Toid	GRID_ExactCut_ID	Location_ExactCut	BoroughName_ExactCut	DotRef	Length (m)	Emissions	Year	Pollutant	...	Artic5Axle	Artic6Axle	PetrolCar	DieselCar	PetrolLgv	DieselLgv	LtBus	Coach
0	5911	4000000027989878	2	External	NonGLA	28440	9.714495	DFT	2013	CO2	...	3.006694	12.549219	18.791658	19.630267	0.279151	11.005820	0.000000	0.744254
1	5911	4000000027989880	2	External	NonGLA	28440	0.000000	DFT	2013	CO2	...	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000
2	5911	4000000027989882	2	External	NonGLA	57226	8.577192	DFT	2013	CO2	...	0.760333	2.446611	19.478135	10.300493	0.120149	7.734197	0.754408	0.868990
3	5911	4000000028014332	2	External	NonGLA	57226	9.347936	DFT	2013	CO2	...	0.823130	2.648621	20.173154	10.553940	0.123945	7.418739	0.820669	0.897038
4	5911	4000000027888882	2	External	NonGLA	28440	0.000000	DFT	2013	CO2	...	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000

	GridId	Toid	GRID_ExactCut_ID	Location_ExactCut	BoroughName_ExactCut	Lts	Length (m)	Emissions	Year	Pollutant	...	Artic6Axle	PetrolCar	DieselCar	PetrolLgv	DieselLgv	DotRef
0	6253	4000000027908919	24	External	NonGLA	18898.0	50.761449	DFT	2013	CO2	...	0.190560	8.761443	4.810774	0.037550	1.735121	NaN
1	6253	4000000027947931	24	External	NonGLA	18895.0	28.592125	DFT	2013	CO2	...	0.000000	0.015535	0.008576	0.000000	0.000000	NaN
2	6253	4000000028013383	24	External	NonGLA	15816.0	5.101391	DFT	2013	CO2	...	0.021509	0.939028	0.518684	0.004055	0.184415	NaN
3	6253	4000000028025820	24	External	NonGLA	15816.0	3.757501	DFT	2013	CO2	...	0.015843	0.691654	0.382044	0.002987	0.135834	NaN
4	6253	4000000028029388	24	External	NonGLA	15816.0	1.624593	DFT	2013	CO2	...	0.006850	0.299044	0.165180	0.001292	0.058729	NaN

	RowID	Year	Toid	GRID_ExactCut_ID	Location_ExactCut	BoroughName_ExactCut	TLRN	MotorwayNumber	AADT Motorcycle	AADT Taxi	...	VKM_Coach	VKM_Rigid2Axle	VKM_Rigid3Axle	VKM_Rigid4Axle	VKM_Artic3Axle	VKM_Artic5Axle	VKM_Artic6Axle	VKM_ElectricCar	VKM_ElectricLgv	VKM_TOTAL
0	1.0	2013.0	4.000000e+15	836.0	Outer	Hillingdon	Other	Other	88.301916	77.112580	...	149.248696	293.680300	55.978941	39.030966	16.191367	10.970609	3.993946	4.335614	1.235289	16605.011414
1	2.0	2013.0	4.000000e+15	2217.0	Outer	Hillingdon	Other	Other	88.301916	77.112580	...	98.338925	193.503902	36.884134	25.717231	10.668379	7.228458	2.631583	2.856706	0.813924	10940.494996
2	3.0	2013.0	4.000000e+15	282.0	External	NonGLA	Other	Other	310.363572	100.322495	...	1657.075319	12950.212101	3011.364039	2861.551314	1710.809301	1966.897025	1647.110606	221.806380	47.635028	796735.125068
3	4.0	2013.0	4.000000e+15	873.0	Outer	Hillingdon	Other	Other	39.473081	144.548284	...	118.008843	9777.985094	2051.227418	1024.275647	470.758531	815.631678	1959.389833	78.775616	15.287825	284144.265992
4	5.0	2013.0	4.000000e+15	2930.0	Outer	Hillingdon	Other	Other	39.473081	144.548284	...	401.216526	33244.027352	6973.937855	3482.419671	1600.524988	2773.054115	6661.700602	267.828056	51.976850	966057.900401

	GridId	Toid	GRID_ExactCut_ID	Location_ExactCut_x	BoroughName_ExactCut_x	Lts	Length (m)_x	Emissions	Year_x	Pollutant	...	VKM_Rigid2Axle	VKM_Rigid3Axle	VKM_Rigid4Axle	VKM_Artic3Axle	VKM_Artic5Axle	VKM_Artic6Axle	VKM_ElectricCar	VKM_ElectricLgv	VKM_TOTAL
0	6253	4000000027908919	24	External	NonGLA	18898.0	50.761449	DFT	2013	CO2	...	2383.880434	229.558857	335.509098	194.242109	247.217230	176.583736	28.990862	5.460174	104036.993985
1	6253	4000000027947931	24	External	NonGLA	18895.0	28.592125	DFT	2013	CO2	...	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.046589	0.000000	140.050860
2	6253	4000000028013383	24	External	NonGLA	15816.0	5.101391	DFT	2013	CO2	...	239.573680	23.070058	33.717777	19.520818	24.844678	17.746199	2.913505	0.548733	10455.442789
3	6253	4000000028025820	24	External	NonGLA	15816.0	3.757501	DFT	2013	CO2	...	176.461358	16.992575	24.835302	14.378333	18.299696	13.071212	2.145983	0.404177	7701.103189
4	6253	4000000028029388	24	External	NonGLA	15816.0	1.624593	DFT	2013	CO2	...	76.294807	7.346907	10.737788	6.216614	7.912054	5.651467	0.927837	0.174750	3329.647849

Table of Contents¶

Importing Libraries¶

Data Preparation¶

Downloading Data¶

Load Data into dataframes¶

Road Emissions Data¶

Traffic Data¶

Join the dataframes¶

Classify columns to labels and targets¶

Visualize the data¶

Derive the target¶

Split data for each type of pollutants¶

Split the data into train and test¶

Feature Selection¶

Model Development¶

Import data if already saved¶

Standardize the data¶

Model Training¶

Linear Regression¶

SVM Regression¶

Hyperparameter Tuning¶

Decision Tree Regression¶

DT Visualisation¶

Hyperparameter Tuning¶

Random Forest Regression¶

Hyperparameter Tuning¶

GradientBoosting Regression¶

Hyperparameter Tuning¶

XGBoost Regression¶

Hyperparameter Tuning¶

Save results¶

Results Summary¶

Pre-processing: feature selection¶

Model Evaluation¶

Output best model results¶

Output pivot table¶

Output best parameters¶

Average results¶

	Pollutant	Feature	Score
94	PM25_Tyre	AADT DLgv	44730.395889
95	PM25_Tyre	AADT Rigid2Axle	43381.929381
96	PM25_Tyre	AADT Rigid3Axle	30302.210535
90	PM25_Tyre	Length (m)_x	27354.984412
93	PM25_Tyre	AADT PLgv	24193.121504

model_name	pollutant	mse	rmse	r2	best_mse	best_rmse	best_r2	best_params
LR	CO2	156343.237076	395.402627	0.622806	156343.237076	395.402627	0.622806	default
LR	NOx	1.333724	1.154870	0.588791	1.333724	1.154870	0.588791	default
LR	PM10_Brake	0.002065	0.045442	0.581906	0.002065	0.045442	0.581906	default
LR	PM10_Exhaust	0.000807	0.028403	0.596459	0.000807	0.028403	0.596459	default
LR	PM10_Resusp	0.007831	0.088491	0.590450	0.007831	0.088491	0.590450	default

	pollutant	best_r2	best_mse	best_rmse
0	CO2	XGB	XGB	XGB
1	NOx	XGB	XGB	XGB
2	PM10_Brake	GBR	GBR	GBR
3	PM10_Exhaust	XGB	XGB	XGB
4	PM10_Resusp	XGB	XGB	XGB
5	PM10_Tyre	XGB	XGB	XGB
6	PM25_Brake	GBR	GBR	GBR
7	PM25_Exhaust	XGB	XGB	XGB
8	PM25_Resusp	GBR	GBR	GBR
9	PM25_Tyre	XGB	XGB	XGB